导入tensorflow时docker停止

Question

我在使用tensorflow构建docker容器时遇到问题。容器的构建很好，但是当它运行脚本'ai_app.py'并到达import tensorflow as tf行时，容器立即停止。它不会向我显示任何错误或其他东西，就像我在docker中使用ctrl+c一样。使用--no-cache标志没有帮助，我得到了相同的结果

下面是我的Dockerfile、运行它的sh脚本以及构建容器时的结果。我正在使用ubuntu 16.04

编辑：我正在添加ai_app.py的第一部分

有人能帮我吗

FROM python:3.7

RUN pip install --upgrade pip

RUN apt-get update
RUN apt-get install -y libsndfile1

COPY requirements.txt /opt/app/requirements.txt
WORKDIR /opt/app
RUN pip install -r requirements.txt
COPY . /opt/app

# Add the keys and set permissions
RUN mkdir /Speech2
WORKDIR /Speech2


CMD ["python", "ai_app.py", "xxxx", "6020", "/Speech"]

#! /bin/bash

docker build -t innovation/speech:latest .

docker run -it \
    --rm \
    --net="host"\
    -v /usr/services/Speech2Text_new/:/Speech2 \
    innovation/speech:latest

Sending build context to Docker daemon  4.096kB
Step 1/11 : FROM python:3.7
 ---> 9337bc3e7477
Step 2/11 : RUN pip install --upgrade pip
 ---> Using cache
 ---> 70d763cdf70c
Step 3/11 : RUN apt-get update
 ---> Using cache
 ---> a11ceed42113
Step 4/11 : RUN apt-get install -y libsndfile1
 ---> Using cache
 ---> 8dbb61916cc6
Step 5/11 : COPY requirements.txt /opt/app/requirements.txt
 ---> Using cache
 ---> c5bee1b79c5e
Step 6/11 : WORKDIR /opt/app
 ---> Using cache
 ---> f6522f756696
Step 7/11 : RUN pip install -r requirements.txt
 ---> Using cache
 ---> a6da3d22e92e
Step 8/11 : COPY . /opt/app
 ---> Using cache
 ---> 89105f62104d
Step 9/11 : RUN mkdir /Speech2
 ---> Using cache
 ---> 77d923bff58d
Step 10/11 : WORKDIR /Speech2
 ---> Using cache
 ---> 89e150986cba
Step 11/11 : CMD ["python", "ai_app.py", "xxxx", "6020", "/Speech"]
 ---> Using cache
 ---> cee3a818ef34
Successfully built cee3a818ef34
Successfully tagged innovation/speech:latest

import sys
import logging
import json
import traceback
import numpy as np
import os
import warnings
import IPython.display as ipd
import librosa
#import torchaudio
#import torchaudio.transforms as T
import tensorflow as tf
from datetime import datetime
from flask import Flask, request, Response, render_template
# from scipy.io import wavfile
# from sklearn.preprocessing import LabelEncoder
#from sklearn.model_selection import train_test_split
#from keras.utils import np_utils
#from keras.layers import Dense, Dropout, Flatten, Conv1D, Input, MaxPooling1D
#from keras.models import Model
#from keras.callbacks import EarlyStopping, ModelCheckpoint
#from keras import backend as K
from tensorflow.keras.models import load_model
from remote_access.fetching import RemoteFiles


class Speech2Text():
    def __init__(self):
        self.classes = None
        self.model = None
        self.labels = None
                 
            ...

Answer 1

我发现，容器停止运行是因为没有启用AVX支持